There is an incomplete understanding of the hormonal, environmental and genetic factors which determine a couple's success with in vitro fertilization (IVF) or gamete intra-fallopian transfer (GIFT). Currently, only age and FSH levels are known predictors of IVF success. Inhibin-A and Activin-A may be mediators of the relationship between elevated FSH levels and poor oocyte quality, and they may be more precise than FSH levels in predicting IVF success. Factors thought to influence early menopause--such as smoking, exposure to galactose, lead or family history of early menopause--may determine why some women have poorer oocyte quality than others. The proposed investigation plans to follow a cohort of 1,600 women who will provide menstrually timed blood and urine samples during a baseline, unstimulated cycle to assess urinary cotinine, galactose-1-phosphate uridyl transferase (GALT) activity and genotype, FSH, LH, estradiol, inhibin-A, activin-A, and lead. Questionnaires from both males and females will assess smoking history, dietary galactose, and general medical and family history. Follicular fluid will be collected during the first IVF/GIFT cycle and tested for cotinine, lead, inhibin, activin and estradiol. Cohort members will be followed for six IVF or GIFT cycles or until pregnancy ending in a live birth. Birth rates and markers of oocyte numbers or quality in relation to the hormonal, genetic, and environmental factors will be assessed using logistic regression.